Making Savings Groups Work for Women - Mango Tree · 2018-07-30 · 2.Walk through the questions with your group a.Prioritize top 2 risks and top 3 metrics 3.Report out in10 minutes

Making Savings Groups Work for Women: Evidence on Gender Transformation

Making Savings Groups Work for WomenEvidence from Randomized Evaluations on the Potential for Gender

Transformation

May 23, 2018Doug Kirke-Smith, IPA Rwanda

Lucía Díaz-Martin, J-PAL at MIT 1


Making Savings Groups Work for Women: Evidence on the Potential for Gender Transformation

Outline1. About IPA and J-PAL2. Impacts of and barriers to saving 3. Savings groups & gender transformation theory of change 4. Evidence from randomized evaluations:

a. Economic outcomes and resilienceb. Women’s control over resources and household decisionsc. Social capital and community participationd. Gender norms

5. What’s next for practice and for research?

2


650+ studies

About IPA

51 countries

650 +studies

3



J-PAL’s mission is to ensure that policy is informed by evidence and research is translated into action

4



J-PAL has more than 900 ongoing and completed projects across 9 sectors in 79 countries

5



Impacts of and barriers to savingEvidence from randomized evaluations

6



Evidence of positive welfare impacts from access to savings

Randomized evaluations from around the world demonstrate the potential for savings to translate to improved welfare: • Increased business investments• Increased income• Increased investments in health or education• Increased food security, food consumption,

and consumption reallocation• Reduced vulnerability to shocks• Increased female decision-making power

Ashraf et al., 2010 (Philippines); Brune et al., 2016 (Malawi); Dupas and Robinson, 2013a (Kenya); Dupas and Robinson, 2013b (Kenya); Karlan et al., 2014 (overview); Karlan and Linden, 2014 (Uganda); Prina, 2015 (Nepal)

7



Savings barriers--exacerbated by gender dynamics-- that savings groups could help address

(Karlan, Ratan, and Zinman 2014; IPA 2017; Schaner 2013; Ashraf et al. 2010; Beaman and Dillon 2017)

Savings Barriers

Proximity

Affordability

Behavioral constraints

Social demands

Trust in formal institutions

Information and knowledge gaps

Gender Dynamics

Social norms related to mobility

Poverty rates among women/woman-headed households

May be exacerbated by time poverty

Control over resources and intra-household bargaining power

May be exacerbated by account requirements biased against women

May be exacerbated by social network asymmetries

8



Savings groups & gender transformation theory of change

9



Theory of change for how savings groups might transform gender relations

ACTIVITIES • Pool savings and receive share-outs • Group formation & meetings

OUTPUTS • Increased savings • Increased contact, discussion, & collaboration between women

OUTCOMES • Improved economic outcomes & resilience

• Increased control over income and assets

• Strengthened group social capital• Increased opportunities for

community leadership & collective action for women

TRANSFORM GENDER RELATIONS

INDIVIDUAL AND HOUSEHOLD COMMUNITY AND COLLECTIVE

10



Collaborative activity1. Turn over the placards on the table2. Walk through the questions with your group

a.Prioritize top 2 risks and top 3 metrics3. Report out in 10 minutes[if you finish early, you can move on to another question!]

11



If you finished your assigned question… move on to another 1. What are the risks that might prevent increased savings from translating into increased control

over income and assets?a. How can we measure increased control over income and assets?

2. What are the risks that might prevent increased contact, discussion, and collaboration between women from translating into strengthened group social capital or community participation and leadership?a. How can we measure strengthened group social capital?b. How can we measure increased opportunities for community participation and leadership?

3. What are the risks that might prevent women’s improved economic outcomes & resilience; increased control over income and assets; and increased social capital, community participation, and leadership from translating into transforming gender relations?a. How can we measure transformed gender relations?

12



Theory of change for how savings groups might transform gender relations

ACTIVITIES

OUTPUTS

OUTCOMES

• Pool savings & receive share-outs • Group formation & meetings

• Increased savings • Increased contact, discussion, & collaboration between women

• Improved economic outcomes & resilience: business investments, consumption, etc.

• Increased control over income & assets: household decisions, spending patterns, life decisions

• Strengthened group social capital

• Increased opportunities for community leadership & collective action

TRANSFORM GENDER RELATIONSGender norms | Bodily integrity | Power structures | Institutions


1

2

3

4

13



1. Economic outcomes and resilience

Evidence from randomized evaluations14


OUTPUTS

OUTCOMES






1

2

3


Evidence of positive impacts on some economic & resilience outcomes for womenLearning brief: some evidence on increased business investments and improved resilience; mixed evidence on additional economic impacts

(Beaman, Karlan, and Thuysbaert 2014; Karlan et al. 2017; Gash 2017)

VSLAs in Ghana, Malawi, and Uganda (CARE & local partners)

Savings for Change in Mali (Oxfam & Freedom from Hunger)

● ↑ savings, business profits (24%), and investments

● No impacts on income or food security

● ↑ savings, livestock holdings (13%), food security (9%), food consumption (3%), and consumption smoothing

● No impacts on overall income or expenditures

15



Can savings groups facilitate specific investments and outcomes?

(Dupas and Robinson 2013)

Study of ROSCAs in Kenya: savings devices designed to increase health investments were effective.• Labeling: soft commitments may suffice

• Increased spending in preventive health by 66-75 % relative to comparison

• The group dynamic led to the largest increases• Increased preventive health spending by 128-

138 % relative to comparison• An emergency fund reduced the likelihood of

being unable to afford treatment by 39 %

16



Can savings groups facilitate gender-specific outcomes?

How can savings groups be modified to reduce the financial barriers to accessing safe childcare delivery services?

• Supplies and transportation to the health care facility are high for many mothers

• Can savings groups offer low-interest loans, pay-outs, or grants for expecting mothers to afford a safer childbirth?

• IPA is currently conducting a study in Zambia to explore this question

17



2. Women’s control over resources and household decisionsEvidence from randomized evaluations

18


OUTPUTS

OUTCOMES






1

2

3


Gender-specific dynamics affect impacts of financial services

Microcredit and business grants research in India, Ghana, Sri Lanka

• Women in households with multiple businesses had lower returns to capital than when they are the only business owner

• Financial resources were often directed to husbands’ businesses

(Bernhardt et al. 2017)19



Evidence from Kenya: intra-household dynamics affected women’s savings behavior, preferences, and needs

Free savings accounts with high withdrawal fees (K-REP)● Entrepreneurs: 87% opened accounts, 47% were

active users 6 months later○ Women: US$41 average deposit ○ Men: US$18 average deposit

• Inaccessibility of funds may have been a factor Free ATM card (Family Bank)• Free ATM card reduced fees for withdrawals• Reduced savings among those with low levels of

household bargaining power

(Schaner 2013, Dupas and Robinson 2013)20



Savings product design: making funds less accessible may work for women with low levels of bargaining power

Commitment savings product in the Philippines (Green Bank):

● Women could not access funds until they reached a goal. ● Access to the accounts:

○ Increased female-oriented durable goods purchased○ Increased women’s scores on a decision-making index

● Largest effects in households where women were married and had below-average bargaining power.

(Ashraf, Karlan, and Yin 2009; Ashraf et al. 2010)21



Mixed evidence on the impact of savings groups on women’s control over income, assets, & household decisions

Learning brief: mixed results on women’s decision-making power

(Beaman, Karlan, and Thuysbaert 2014; Karlan et al. 2017; IPA, BARA 2013; Gash 2017; Dupas and Robinson 2013)

VSLAs in Ghana, Malawi, and Uganda

Savings for Change in Mali

↑ control over business decisions (8%), food expenses (5%), & education expenses (4%)

No impact, perhaps due to the patriarchal customs

Boxes for ROSCA members in Kenya 43% of box users: helped say no to spouses.

22



Can we design for changes in household decision-making more explicitly?

• Can engagement with men increase impacts on household decision-making?

• IPA is currently conducting studies in Tanzania and Nicaragua examining the impact of training on gender equitable household decision-making on the effects of mobile money interventions.

23



Lessons from the audience: What have been some of your own innovations in increasing economic, resilience, or household control outcomes for women?

24



3. Social capital and community participation

Evidence from randomized evaluations

25


OUTPUTS

OUTCOMES






1

2

3


What is the role of the group in building social capital and community participation?Group composition and meeting frequencies matter• Women in mixed-gender groups in Benin had lower social capital than

those in women-only groups• More frequent group lending meetings among women in India led to

higher levels of social interaction that extended beyond the meetings Peer support may play a role in driving positive impacts• Women borrowers who attended a training with a friend in India had

higher business sales compared to women who attended alone, especially among women subject to restrictive norms

(Field, Jayachandran, and Pande 2010; Karlan, Thuysbaert, and Gray 2017; Feigenberg et al. 2014) 26



Mixed evidence on the impact of savings groups on social capital and community participationLearning brief: positive impacts on group solidarity, but mixed results on community leadership and activism and limited impacts on social networks

(Beaman, Karlan, and Thuysbaert 2014; Karlan et al. 2017; IPA, BARA 2013; Gash 2017)

VSLAs in Ghana, Malawi, and Uganda

Savings for Change in Mali

● ↑ 5 % likelihood of attending community meeting ● No impacts on community membership or speaking

to community leaders

● No effect on women’s social capital (social integration and community action)

● Qualitative research: increased perceptions

27



Open questions for future research and practice related to groups and social capital

Design considerations• What are the mechanisms behind increasing social capital via group-

based financial interventions? • How can digital products be designed in a way that preserves or enhances

the group component?Leveraging groups for community-level impacts• How can savings groups be leveraged to encourage collective action

toward shared, non-economic goals and for women to take community leadership positions?

Measurement• What are the best metrics for capturing social capital?

28



4. Gender norms

Evidence from randomized evaluations


4

29


OUTCOMES • Improved economic outcomes & resilience: business investments, consumption, etc.




1

2

3


Can we design for gender norms change explicitly? Preliminary evidence on bodily integrity● IRC VSLA intervention promoting gender equitable norms with women and their

partners in Cote d’Ivoire reduced intimate partner violence:

● Engaging men may be crucial: similar program in Peru targeted to women microfinance clients alone found no reduction in reported violence or changes in social norms

Reported economic abuse

Reported intimate partner violence

34.6 % 21.2 %

14.8 % 7.5 %

VSLA only VSLA + gender dialogue (high participation)

(Gupta et al. 2013; Agüero and Frisancho 2018)

In the last year:

30



Empowerment and Livelihood for Adolescents (ELA) by BRAC (Uganda and Tanzania)

Results in Uganda, but not Tanzania

Hard and soft skills in combination were effective and implementation details matter

Hard & soft skills after

school program for

youth

↓ 44 % of girls having sex unwillingly

↑ 72% in income generating activities; ↑ consumption

Shifted attitudes and aspirations about employment, marriage and childbearing

(Bandiera et al. 2015; Buehren et al. 2017) 31



Breakthrough classroom based program in India

Behavior change was possibly easier for boys than girls, and parental attitudes are a factor

Secondary school-based

gender attitude

campaign

↑ gender attitude to be more associated with gender equality

↑ gender behavior to be more equitable, more so for boys

Parent attitudes influence child attitudes

(Jayachandra, Jain, and Dhar 2016) 32



Kishoree Kontha (Bangladesh Development Society & Save the Children) empowerment program in Bangladesh

Empowerment program didn’t affect child marriage, but did have impacts on education and income-generating activities

Empowerment training

program

↑ being in school by 7%

↑ ever having an income generating activity by 51%

No impacts on child marriage

(Buchmann et al. 2017) 33



Girls Arise! negotiation training program in Zambia

Negotiation program helped girls ask for more from parents compared to group after-school meetings with no training

Girls Arise! primary school

negotiation program

↑ girls’ welfare (index measuring pregnancy and educational outcomes)

Strongest effects for girls who were on the margin of continuing school

Compared to Safe Space program which had no impacts

(Ashraf et al. 2017) 34



Preliminary takeaways on programming to change gender norms

● Existing power structures which vary by context can limit impacts

● Targeting women & girls may work better in domains where they have more choice

● Working with men or other decision-makers may be helpful in other domains

● Program piloting and ongoing research may help determine promising pathways

35



Open questions for future research and practice on gender transformation

Who to target?• What other actors need to be involved to achieve the desired impact?

When to target?• How can we target interventions before girls get married to impact

adolescents’ ability to negotiate for resources for education and health?

What else can we do with savings groups? • What other kinds of gender norms programming might be impactful in a

savings group context?

36



So what’s next for practice and for research?

37



Summary of implications for future practice: savings groups & gender transformation

• Timing to increase economic outcomes at key points• Labeling funds as a soft commitment device for purchases

relevant to women or to enhance ability to exert control over resources

• Male engagement and community sensitization to address areas in which women aren’t the key decision-makers

• Targeting interventions for young women such as negotiation and soft skills to support empowerment outcomes

• Incorporating additional gender norms programming to explicitly address social issues

38



Summary of implications for future research: savings groups & gender transformation

• Group dynamics: specific mechanisms, digital group design• Community and collective impact: identifying the best pathways• Gender norms programming with savings groups: additional

interventions and approaches• Measurement: What are the best metrics for capturing social capital

and women’s decision-making?• Formal vs. informal: Is promoting informal savings groups the right

path, or is more formalized institutional access key for improving women's power?

39



40


ACTIVITIES

OUTPUTS

OUTCOMES

• Pool savings & receive share-outs • Group formation & meetings








1

2

3

4

Back to the theory of change


Lessons from the audience and applying the evidence (use the post-its!):

1. What have you done when it comes to gender transformative work and how did it go?

2. What are you thinking about doing next?• Timing to increase economic outcomes at key points• Labeling funds as a soft commitment device for purchases relevant to

women or to enhance ability to exert control over resources• Engaging men and community sensitization to address areas in which

women aren’t the key decision-makers• Targeting interventions for young women such as negotiation and soft

skills to support empowerment outcomes• Incorporating additional gender norms programming to explicitly

address social issues41



APPENDIX

42


Evidence from Nepal: women have diverse needs and preferences

Free formal savings accounts for women in Nepal • 84 percent of HHs opened

accounts• 80 percent were active users one

year later• Spent 20 percent more on

education, 15 percent more on meat and fish, and were better able to cope with health shocks

(Prina 2013)43


Documents

Making Savings Groups Work for Women - Mango Tree · 2018-07-30 · 2.Walk through the questions with your group a.Prioritize top 2 risks and top 3 metrics 3.Report out in10 minutes